Modulation of the c-Jun N-terminal kinase activity
in the embryonic heart in response to anoxia-reoxygenation:
involvement of the Ca
2+and mitoK
ATPchannels
Alexandre SarreÆ Ste´phany Gardier Æ Fabienne MaurerÆ Christophe Bonny Æ Eric Raddatz
Received: 24 December 2007 / Accepted: 28 March 2008 / Published online: 17 April 2008 Ó Springer Science+Business Media, LLC. 2008
Abstract Whether the response of the fetal heart to ischemia-reperfusion is associated with activation of the c-Jun N-terminal kinase (JNK) pathway is not known. In contrast, involvement of the sarcolemmal L-type Ca2+ channel (LCC) and the mitochondrial KATP (mitoKATP) channel has been established. This work aimed at investi-gating the profile of JNK activity during anoxia-reoxygenation and its modulation by LCC and mitoKATP channel. Hearts isolated from 4-day-old chick embryos were submitted to anoxia (30 min) and reoxygenation (60 min). Using the kinase assay method, the profile of JNK activity in the ventricle was determined every 10 min throughout anoxia-reoxygenation. Effects on JNK activity of the LCC blocker verapamil (10 nM), the mitoKATP channel opener diazoxide (50 lM) and the blocker 5-hy-droxydecanoate (5-HD, 500 lM), the mitochondrial Ca2+ uniporter (MCU) inhibitor Ru360 (10 lM), and the anti-oxidant N-(2-mercaptopropionyl) glycine (MPG, 1 mM) were determined. In untreated hearts, JNK activity was increased by 40% during anoxia and peaked fivefold rela-tive to basal level after 30–40 min reoxygenation. This peak value was reduced by half by diazoxide and was
tripled by 5-HD. Furthermore, the 5-HD-mediated stimu-lation of JNK activity during reoxygenation was abolished by diazoxide, verapamil or Ru360. MPG had no effect on JNK activity, whatever the conditions. None of the tested pharmacological agents altered JNK activity under basal normoxic conditions. Thus, in the embryonic heart, JNK activity exhibits a characteristic pattern during anoxia and reoxygenation and the respective open-state of LCC, MCU and mitoKATPchannel can be a major determinant of JNK activity in a ROS-independent manner.
Keywords JNK mitoKATPchannel Calcium channel Anoxia-reoxygenation Embryo
Introduction
The signaling pathways involved in the response of the fetal heart to inadequate oxygenation, resulting from transient maternal hypoxemia, reduction in uterine or umbilical blood flow, remain to be explored. We have previously characterized in detail the electrical and con-tractile disturbances induced by anoxia and reoxygenation in the embryonic heart model [1, 2] and found that a moderate inhibition of the sarcolemmal L-type calcium channel (LCC) [3] or activation of the mitochondrial ATP-sensitive potassium (mitoKATP) channel can improve postanoxic recovery [4]. Opening of the mitoKATP chan-nel is also involved in ischemic preconditioning of isolated embryonic ventricular myocytes [5] and adult heart [6].
Several studies have underlined the key role played by the mitogen-activated protein kinases (MAPKs) pathways [7] in myocardial ischemia and reperfusion, particularly the stress-activated c-Jun N-terminal kinase (JNK) [8, 9]. In
A. Sarre and S. Gardier contributed equally to this work. A. Sarre S. Gardier E. Raddatz (&)
Department of Physiology, Faculty of Biology and Medicine, University of Lausanne, 7 rue du Bugnon, 1005 Lausanne, Switzerland
e-mail: eric.raddatz@unil.ch A. Sarre
e-mail: alexandre.sarre@medecine.unige.ch F. Maurer C. Bonny
Service of Medical Genetics, Centre Hospitalier Universitaire Vaudois and University of Lausanne, Lausanne, Switzerland DOI 10.1007/s11010-008-9750-4
the ventricle of the anoxic-reoxygenated embryonic chick heart, activity of the extracellular signal-regulated kinase (ERK) is not significantly altered and the profile of the p38 MAPK phosphorylation is not affected by opening of the mitoKATP channel [10]. However, information regarding the JNK signaling pathway in the hypoxic fetal heart is lacking and a better understanding is especially required in the context of recent advances in developmental cardiology [11], fetal cardiac surgery [12] and research dealing with intrauterine programming [13]. Furthermore, as Ca2+is one of the second messengers capable of modulating JNK activity [14], the cytosolic and mitochondrial Ca2+ over-load induced by anoxia-reoxygenation in embryonic cardiomyocytes [15] could interfere with the JNK signaling pathway.
The aim of this work was to establish the profile of JNK activity in the ventricle of the embryonic heart during anoxia and reoxygenation and to investigate the possible link between JNK activity and the state of activation of the Ca2+and mitoKATPchannels. The results suggest that LCC and mitochondrial Ca2+and KATPchannels are involved in the modulation of JNK activity in the embryonic ventricle submitted to anoxia-reoxygenation.
Materials and methods
Reagents
Dimethylsulfoxide (DMSO), mitoKATP channel opener diazoxide and blocker 5-hydroxydecanoate (5-HD), radical scavenger N-(2-mercaptopropionyl)glycine (MPG), were purchased from Sigma (Sigma–Aldrich, Buchs, Switzer-land). L-Type Ca2+channel inhibitor verapamil (IsoptinÒ) was from Abbott and mitochondrial Ca2+uniporter (MCU) inhibitor Ru360 was purchased from Calbiochem (JURO Supply, Lucerne, Switzerland).
[c-33P]ATP was from Amersham Biosciences and inhibitors of proteases from Roche Biosciences.
Preparation and in vitro mounting of the heart
Fertilized eggs from Lohman Brown hens were incubated during 96 h at 38°C and 95% relative humidity to obtain stage 24 HH embryo (according to Hamburger and Ham-ilton [16]). Spontaneously beating hearts were carefully excised and placed in the culture compartment of an air-tight stainless steel chamber. The chamber was equipped with two windows for observation and maintained under controlled conditions on the thermostabilized stage (37.5°C) of an inverted microscope (IMT2 Olympus, Tokyo, Japan) as previously detailed [4]. Briefly, the cul-ture compartment (300 ll) was separated from the gas
compartment by a 15 lm transparent and gas-permeable silicone membrane (RTV 141, Rhoˆne-Poulenc, Lyon, France). Thus, pO2 at the tissue level could be strictly controlled and rapidly modified (within less than 5 s) by flushing high-grade gas of selected composition through the gas compartment. At this developmental stage, the heart lacks vascularization and the myocardial oxygen requirement is met exclusively by diffusion.
The standard HCO3/CO2 buffered medium was com-posed of (in mmol/l): 99.25 NaCl; 0.3 NaH2PO4; 10 NaHCO3; 4 KCl; 0.79 MgCl2; 0.75 CaCl2; 8 D+glucose. This culture medium was equilibrated in the chamber with 2.31% CO2in air (normoxia and reoxygenation) or in N2 (anoxia) yielding a pH of 7.4. All reagents were diluted in this medium containing 0.5% DMSO (vehicle).
Anoxia-reoxygenation protocol
After 45 min of in vitro stabilization under normoxia at 37.5°C (stab), the hearts were submitted to strict anoxia during 30 min and then reoxygenated during 60 min. The pharmacological agents were present throughout anoxia-reoxygenation. The hearts were collected every 10 min and the ventricle was carefully dissected on ice and stored at -80°C for subsequent determinations. As control, in a separate set of experiments, hearts were maintained under steady normoxia for 60 and 90 min after stabilization, corresponding to the time points of 30 and 60 min of postanoxic reoxygenation, respectively. Ventricles of these untreated hearts were also dissected and stored at -80°C.
Kinase assay
Ventricular JNK activity was determined using a published method [17] with minor modifications. Ventricles were homogenized in ice-cold lysis buffer (in mmol/l: 20 Tris-acetate (pH 7), 270 sucrose, 1 EGTA, 1 EDTA, 50 NaF, 10 b-glycerophosphate, 10 dithiothreitol (DTT), 10 4-nitro-phenyl phosphate disodium salt hexahydrate (PNPP), 1% Triton X-100 and inhibitors of proteases). Insoluble mate-rial was removed by a 5 min centrifugation at 10,000g and protein contents were measured by the method of Lowry [18] with bovine serum albumin as standard.
Soluble ventricular protein extract (30 lg) were incubated for 3 h at 4°C in the presence of 1 lg GST-c-Jun(1–219) bound to gluthatione-agarose beads as both JNK-specific ligand and substrate. The beads were washed three times in washing buffer (same as lysis buffer but with 0.1% Triton X-100) and twice in kinase buffer (in mmol/l, 20 HEPES pH 7.5, 10 MgCl2, 20 b-glycerophosphate, 10 DTT, 10 PNPP and inhibitors of proteases). Kinase reaction was carried out for 30 min at 30°C in 20 ll of kinase buffer containing 5 lCi
[c-33P]ATP. Reaction products were resolved by 12% SDS-polyacrylamide gel electrophoresis (SDS-PAGE), gels were dried, and phosphorylation signals were visualized by autoradiography, quantified by PhosphoImager (Quantity-one 1.4.0, Biorad) and expressed as fold increase relative to the respective preanoxic value (stab).
Statistical analysis
All values are reported as mean ± standard error of the mean (S.E.). The significance of any difference between the groups was assessed using Student t-test. The statistical significance was defined by a value of P \ 0.05.
Results
Profile of JNK activity during anoxia-reoxygenation
In control conditions (vehicle), JNK activity increased by 40% after 10–20 min anoxia (P \ 0.05). During reoxy-genation, JNK activity progressively increased, peaking after 30–40 min and further declined (Fig.1). This cul-mination of JNK activity was specifically related to reoxygenation, since it was not attributable to the condi-tions and the duration (135 min) of culture in the chamber (Fig.2).
Modulation of JNK activity by the opening state of the mitoKATP channel
The mitoKATP channel opener diazoxide (50 lmol/l) decreased JNK activity after 30 min anoxia and throughout reoxygenation (Fig.1). As the diazoxide-induced inhibition was the strongest at the peak of JNK activity (P \ 0.02), we selected this time point to phar-macologically explore the possible mechanisms linking the open state of the mitoKATP channel with the JNK activity. Under preanoxic conditions, however, none of
0 1 2 3 4 5 6 fold increase vehicle diazoxide 10 20 30 10 20 30 40 50 60 min stab ANOXIA REOXYGENATION * # * * * * * * * * * * # # # # # 1512 4 4 4 4 4 4 4 4 4 4 159 109 4 4 4 4 ANOXIA REOXYGENATION stab P-c-Jun P-c-Jun vehicle diazoxide 10 20 30 10 20 30 40 50 60 min
Fig. 1 Profile of JNK activity in the embryonic ventricle during anoxia-reoxygenation. Activation of the mitoKATP channel by diazoxide decreased JNK activity. Upper panels show representative
autoradiogram of JNK activity during anoxia and
reoxygenation. Fold increase: JNK activity is given relative to the preanoxic stab value of vehicle. Stab: 45 min preanoxic stabilization; mean ± S.E. of number of determinations indicated in columns; *P \ 0.05 vs. stab;#P\ 0.05 vs. vehicle 0 0.5 1 1.5 2 fold increase stab stab + 60 min normoxia stab + 90 min normoxia
Fig. 2 Stability of JNK activity under normoxia. In vitro culture under steady normoxia did not affect ventricular JNK activity assessed by kinase assay. Fold increase: JNK activity is given relative to the mean of stab values. Stab: 45 min preanoxic stabilization; stab + 60 min and stab + 90 min correspond to 30 and 60 min of reoxygenation in the anoxia-reoxygenation protocol, respectively (see Fig.1). Mean ± S.E.; n = 6–8 determinations of JNK activity
the reagents, alone or in combination, affected JNK activity relative to vehicle (Fig.3a) or disturbed the regular contractile activity of the isolated hearts (not shown). After 30–40 min reoxygenation, inhibition of JNK activity by diazoxide was suppressed by the mitoKATP channel blocker 5-HD (500 lmol/l) and, importantly, 5-HD alone tripled JNK activity with respect to vehicle (P \ 0.01) (Fig.3b). The radical scavenger MPG (1 mmol/l), known otherwise to abolish the diaz-oxide-induced ROS production and cardioprotection at reoxygenation [4], affected neither JNK activity nor the diazoxide-induced JNK inhibition.
Modulation of JNK activity by sarcolemmal (LCC) and mitochondrial (MCU) Ca2+channels
Relative to vehicle, LCC inhibitor verapamil (10 nmol/l) decreased JNK activity by 55%, whereas the mitochondrial Ca2+uniporter inhibitor Ru360 (10 lmol/l) had no signif-icant effect (Fig.3b). However, the 5-HD-mediated JNK activity during reoxygenation was abolished by verapamil and also by Ru360. These observations indicate that Ca2+ entry is a prerequisite for JNK stimulation and that MCU is involved in JNK activation induced by the mitoKATP channel blocker 5-HD.
Discussion
To the best of our knowledge, this is the first time that JNK activity is explored in the embryonic myocardium sub-mitted to an anoxic episode. Our main findings indicate that JNK activity in the ventricle of the isolated embryonic heart (1) is increased by anoxia and reoxygenation, (2) is modulated by the open-state of the mitoKATPchannel, and (3) is dependent on Ca2+flux through both LCC and MCU. In the adult heart the effects of ischemia on JNK acti-vation remain controversial whereas all studies show an enhanced JNK activity during reperfusion [8, 9, 19, 20]. Our data indicate that JNK pathway contributes to the short-term response of the heart to oxygen lack and rein-troduction also during early embryogenesis, although the metabolic consequences of anoxia-reoxygenation differ to a certain extent from those of ischemia-reperfusion.
Although 5-HD has been shown to abolish ROS pro-duction induced by the mitoKATPchannel opener diazoxide during reoxygenation [4], it markedly increased JNK activity (Fig.3b). Furthermore, the membrane permeable antioxidant MPG which significantly reduces ROS pro-duction at reoxygenation [4] did not suppress JNK activation. Taken together, these observations clearly indicate that endogenous ROS are not prerequisite for JNK
PREANOXIC STABILIZATION 0 2 6 2 1 11 9 5 9 6 6 5 6 fold increase A 30-40 min REOXYGENATION 0 2 4 6 8 10 12 14
vehicle DIAZ DIAZ + 5-HD DIAZ + MPG MPG verap D H -5 5-HD + verap Ru 5-HD + Ru * * * * # § § 6 6 8 4 1 8 4 1 8 1 5 2 fold increase 6 6 B ¶ vehicle verap 5-HD + verap P-c-Jun 5-HD + Ru 5-HD Ru
Fig. 3 Pharmacological modulation of JNK activity in the embryonic ventricle during preanoxic stabilization (a) and after 30–40 min reoxygenation (b). JNK activity was dependent on the open-state of the mitoKATP channel through Ca2+-dependent mechanisms during reoxygenation exclusively. In panel (a), JNK activity is given relative to the value of vehicle. In panel (b), JNK activity is given relative to
the respective preanoxic stabilization value shown in panel (a). Insets show representative autoradiograms of JNK activity during reoxy-genation in relation to Ca2+ handling. DIAZ: diazoxide; verap: verapamil; Ru: Ru360; Mean ± S.E. of number of determinations indicated in columns; *P \ 0.05 vs. vehicle; #P\ 0.05 vs. DIAZ; §P\ 0.05 vs. 5-HD;}P
activation in the embryonic myocardium by contrast with neonatal cardiomyocytes [21] and adult heart [8]. Such a developmental discrepancy implies that JNK pathway can be modulated differently in prenatal and postnatal myo-cardium since, for example, the physiological oxygen level, the oxidative metabolism and the capacity to generate ROS are lower in embryonic tissues [22–24]. The facts that diazoxide improves recovery of atrio-ventricular conduc-tion and E-C coupling during the first 20 min of reoxygenation [4], and reduces JNK activity from the end of anoxia onward (Fig.1), suggest that this protection might be indirectly related to a reduction of JNK activity. It has been shown that after 30 min reoxygenation, that is, when JNK activity reached its highest value (Fig.1), Ca2+ uptake is maximally increased in embryonic ven-tricular cells [15] and contractility of the ventricle is transiently above its basal level [25], reflecting a rise of intracellular Ca2+. The present data support the concept that extracellular Ca2+entry through LCC is a prerequisite for JNK activation in the reoxygenated embryonic myo-cardium (Fig.3b) alike in neonatal [21] and adult cardiomyocytes [8].
Subcellular fractionation studies have shown that JNK can also be localized within or associated with mitochon-drial structures [26, 27]. The facts that mitochondria are capable of taking up some of the cytosolic Ca2+through the MCU in case of Ca2+overload [28,29] and that mitoKATP channel opening can decrease the mitochondrial inner membrane potential (Wm) and Ca2+content in the matrix during hypoxia [30] could partly explain that MCU inhi-bition suppressed the 5-HD-induced JNK activation. Our results, indeed, show that blocking moderately Ca2+entry through sarcolemmal LCC or blocking Ca2+ flux into mitochondrion through MCU reduces JNK activity during reoxygenation, especially when mitoKATP channel are closed by 5-HD.
Although Wm and mitochondrial Ca2+ concentration have not been measured because of the thickness of the tissue and interferences due to contractions, our findings support the hypothesis that during reoxygenation Ca2+ entry through LCC and/or influx through MCU can activate cytosolic JNK and/or JNK associated with mitochondria (see model proposed in Fig.4).
When mitoKATPchannels are opened by diazoxide, Wm is known to drop [30,31], thereby limiting mitochondrial Ca2+entry through MCU and consequently Ca2+overload, reducing JNK activation. Conversely, when mitoKATP channels are blocked by 5-HD, Wm should be maintained and act as a driving force for Ca2+transport through MCU, which also could contribute to activate JNK.
Thus, in the embryonic heart, JNK activity exhibits a characteristic pattern during anoxia and reoxygenation and the respective open-state of LCC, MCU and mitoKATP
channels can be a major determinant of JNK activity in a ROS-independent manner. This work provides a first step in understanding the regulation of the JNK signaling pathway in the fetal heart transiently exposed to hypoxia. In partic-ular, the cellular targets as well as the long-term functional consequences of an acute activation of this pathway during cardiogenesis deserve further investigations.
Acknowledgements We thank Anne-Catherine Thomas for her skilful technical assistance. This work was supported by the Swiss National Science Foundation n°3100A0-105901.
References
1. Sarre A, Maury P, Kucera P et al (2006) Arrhythmogenesis in the developing heart during anoxia-reoxygenation and hypothermia-rewarming: an in vitro model. J Cardiovasc Electrophysiol 17:1350–1359
2. Sedmera D, Kucera P, Raddatz E (2002) Developmental changes in cardiac recovery from anoxia-reoxygenation. Am J Physiol Regul Integr Comp Physiol 283:R379–R388
MCU i.m. o.m. MITOCHONDRION S.M. [Ca2+] c JNK? JNK [Ca2+] m
- +
Ψm verapamil LCC Ru360 mitoKATP DIAZ 5-HD ROS MPG I II III IV Ψm ↑or↓ ?Fig. 4 Schematic representation based on the present findings and illustrating the possible modulation of JNK activity by open-state of L-type Ca2+channel (LCC), mitochondrial Ca2+uniporter (MCU) and mitoKATPchannel during reoxygenation of the embryonic ventricle. Inactivation of LCC (verapamil) decreased the reoxygenation-induced peak of JNK activity. Inhibition of MCU (Ru360) or LCC suppressed the 5-HD-induced JNK activity (see Fig.3). Opening (DIAZ) and blockade (5-HD) of the mitoKATP channel could respectively decrease and increase Wm, [Ca2+]m and JNK activity. Intracellular antioxidant (MPG) had no effect on JNK activity. S.M.: sarcolemmal membrane; i.m. and o.m.: inner and outer mitochondrial membrane; Wm: i.m. potential; [Ca2+]c and [Ca2+]m: cytosolic and mitochondrial Ca2+ concentrations; MCU: mitochondrial Ca2+ uniporter; I–IV: respiratory chain complexes; ROS: reactive oxygen species
3. Tenthorey D, de Ribaupierre Y, Kucera P et al (1998) Effects of verapamil and ryanodine on activity of the embryonic chick heart during anoxia and reoxygenation. J Cardiovasc Pharmacol 31:195–202
4. Sarre A, Lange N, Kucera P et al (2005) mitoKATP channel activation in the postanoxic developing heart protects E-C cou-pling via NO-, ROS-, and PKC-dependent pathways. Am J Physiol Heart Circ Physiol 288:H1611–H1619
5. Lebuffe G, Schumacker PT, Shao ZH et al (2003) ROS and NO trigger early preconditioning: relationship to mitochondrial KATP channel. Am J Physiol Heart Circ Physiol 284:H299– H308
6. Garlid KD, Paucek P, Yarov-Yarovoy V et al (1997) Cardio-protective effect of diazoxide and its interaction with mitochondrial ATP-sensitive K+ channels. Possible mechanism of cardioprotection. Circ Res 81:1072–1082
7. Schulz R, Cohen MV, Behrends M et al (2001) Signal trans-duction of ischemic preconditioning. Cardiovasc Res 52:181–198 8. Knight RJ, Buxton DB (1996) Stimulation of c-Jun kinase and mitogen-activated protein kinase by ischemia and reperfusion in the perfused rat heart. Biochem Biophys Res Commun 218:83–88 9. Omura T, Yoshiyama M, Shimada T et al (1999) Activation of mitogen-activated protein kinases in in vivo ischemia/reperfused myocardium in rats. J Mol Cell Cardiol 31:1269–1279
10. Gardier S, Sarre A, Thomas AC et al (2006) Activation of the mitoKATP channel differently modulates ERK, JNK and p38 MAPK in the ventricle of the anoxic-reoxygenated developing heart. Arch Mal Coeur Vaisseaux 99:376 (abstract)
11. Strasburger JF (2005) Prenatal diagnosis of fetal arrhythmias. Clin Perinatol 32:891–912
12. Kumar S, O’Brien A (2004) Recent developments in fetal med-icine. BMJ 328:1002–1006
13. Fowden AL, Giussani DA, Forhead AJ (2006) Intrauterine pro-gramming of physiological systems: causes and consequences. Physiology (Bethesda) 21:29–37
14. Schaub MC, Hefti MA, Zaugg M (2006) Integration of calcium with the signaling network in cardiac myocytes. J Mol Cell Cardiol 41:183–214
15. Murphy JG, Smith TW, Marsh JD (1988) Mechanisms of reoxygenation-induced calcium overload in cultured chick embryo heart cells. Am J Physiol 254:H1133–H1141
16. Hamburger V, Hamilton H (1951) A series of normal stages in the development of the chick embryo. J Morphol 88:49–92 17. Larsen CM, Wadt KA, Juhl LF et al (1998)
Interleukin-1beta-induced rat pancreatic islet nitric oxide synthesis requires both the p38 and extracellular signal-regulated kinase 1/2 mitogen-activated protein kinases. J Biol Chem 273:15294–15300
18. Lowry O, Rosebrough N, Farr A et al (1951) Protein measure-ment with the Folin phenol reagent. J Biol Chem 193:265–275 19. Hreniuk D, Garay M, Gaarde W et al (2001) Inhibition of c-Jun
N-terminal kinase 1, but not c-Jun N-terminal kinase 2, sup-presses apoptosis induced by ischemia/reoxygenation in rat cardiac myocytes. Mol Pharmacol 59:867–874
20. Ren J, Zhang S, Kovacs A et al (2005) Role of p38alpha MAPK in cardiac apoptosis and remodeling after myocardial infarction. J Mol Cell Cardiol 38:617–623
21. Dougherty CJ, Kubasiak LA, Frazier DP et al (2004) Mito-chondrial signals initiate the activation of c-Jun N-terminal kinase (JNK) by hypoxia-reoxygenation. Faseb J 18:1060–1070 22. Allen RG (1991) Oxygen-reactive species and antioxidant
responses during development: the metabolic paradox of cellular differentiation. Proc Soc Exp Biol Med 196:117–129
23. Lebovitz RM, Zhang H, Vogel H et al (1996) Neurodegeneration, myocardial injury, and perinatal death in mitochondrial super-oxide dismutase-deficient mice. Proc Natl Acad Sci USA 93:9782–9787
24. Romano R, Rochat AC, Kucera P et al (2001) Oxidative and glycogenolytic capacities within the developing chick heart. Pe-diatr Res 49:363–372
25. Rosa A, Maury JP, Terrand J et al (2003) Ectopic pacing at physiological rate improves postanoxic recovery of the devel-oping heart. Am J Physiol Heart Circ Physiol 284:H2384–H2392 26. Horbinski C, Chu CT (2005) Kinase signaling cascades in the mitochondrion: a matter of life or death. Free Radic Biol Med 38:2–11
27. Wiltshire C, Gillespie DA, May GH (2004) Sab (SH3BP5), a novel mitochondria-localized JNK-interacting protein. Biochem Soc Trans 32:1075–1077
28. O’Rourke B, Cortassa S, Aon MA (2005) Mitochondrial ion channels: gatekeepers of life and death. Physiology (Bethesda) 20:303–315
29. Anderson CD, Pierce J, Nicoud I et al (2005) Modulation of mitochondrial calcium management attenuates hepatic warm ischemia-reperfusion injury. Liver Transpl 11:663–668 30. Kim MY, Kim MJ, Yoon IS et al (2006) Diazoxide acts more as a
PKC-epsilon activator, and indirectly activates the mitochondrial K(ATP) channel conferring cardioprotection against hypoxic injury. Br J Pharmacol 149:1059–1070
31. Comelli M, Metelli G, Mavelli I (2007) Downmodulation of mitochondrial F0F1 ATP synthase by diazoxide in cardiac myoblasts: a dual effect of the drug. Am J Physiol Heart Circ Physiol 292:H820–H829
